On the Evaluation and Comparison of Taggers: The Effect of Noise in Testing Corpora

dc.creatorPadro, L.
dc.creatorMarquez, L.
dc.date1998-09-28
dc.date.accessioned2026-07-07T03:23:41Z
dc.date.available2026-07-07T03:23:41Z
dc.descriptionThis paper addresses the issue of {\sc pos} tagger evaluation. Such evaluation is usually performed by comparing the tagger output with a reference test corpus, which is assumed to be error-free. Currently used corpora contain noise which causes the obtained performance to be a distortion of the real value. We analyze to what extent this distortion may invalidate the comparison between taggers or the measure of the improvement given by a new system. The main conclusion is that a more rigorous testing experimentation setting/designing is needed to reliably evaluate and compare tagger accuracies.
dc.descriptionAppears in proceedings of joint COLING-ACL 1998, Montreal, Canada
dc.identifierhttps://arxiv.org/abs/cs/9809112
dc.identifierhttp://arxiv.org/abs/cs/9809112
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/33040
dc.subjectComputation and Language
dc.subjectI.2.7
dc.titleOn the Evaluation and Comparison of Taggers: The Effect of Noise in Testing Corpora
dc.typetext

Files

Collections